METHOD AND SYSTEM FOR PREDICTING EQUIPMENT FAILURE

• US 20160274551A1
• Filed: 05/20/2015
• Published: 09/22/2016
• Est. Priority Date: 03/18/2015
• Status: Active Grant

• Current Assignee: Accenture Global Services Limited (Accenture PLC)
• Original Assignee: Accenture Global Services Limited (Accenture PLC)
• Inventors: Mishra, Rabinarayan, Jalihal, Akshay, Verma, Sandeep, Vegesna, Ram Gopal Varma
• Granted Patent: US 11,042,128 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G05B 15/02 electric
• CPC Class Codes: G05B 9/02 electric

Abstract

A method for predicting equipment failure includes receiving parameter sample values associated with parameters of benchmark equipment and operational status information associated with the benchmark equipment. The parameter sample values and operational status information are periodically acquired. A model is generated for relating one or more of the parameters to benchmark equipment failure. For each parameter, a threshold value at which an output of the model indicates benchmark equipment failure is determined. Next, parameters of an equipment under test having parameter sample values that match the determined threshold values are determined. For each determined parameter, benchmark equipment having parameter sample values that match the parameter sample values of the equipment under test that match the determine threshold values are determined. Survivability data for the equipment under test is generated based on survivability data associated with the determined benchmark equipment. The generated survivability data is communicated to a user interface.

Claims

1. A method for predicting equipment failure, the method comprising:

receiving parameter sample values associated with parameters of benchmark equipment and operational status information associated with the benchmark equipment, wherein the parameter sample values and operational status information are periodically acquired;

generating a model that relates one or more of the parameters to benchmark equipment failure;

for each of the one or more parameters, determining a threshold value at which an output of the model indicates benchmark equipment failure;

determining parameters of an equipment under test for which at least some parameter sample values associated with the parameters match the determined threshold values;

for each determined parameter, determining benchmark equipment having parameter sample values that match the parameter sample values of the equipment under test;

generating survivability data for the equipment under test based on survivability data associated with the determined benchmark equipment; and

communicating the generated survivability data to a user interface.

2. The method according to claim 1, further comprising:

correlating the parameter sample values associated with parameters of benchmark equipment with the operational status information of benchmark equipment to determine one or more parameters having a greatest relationship to the operational status of the benchmark equipment; and

wherein the one or more parameters related to the benchmark equipment failure by the model correspond to those parameters having a correlation higher than a predetermined threshold.

3. The method according to claim 1, wherein the generated model corresponds to a linear regression model.

4. The method according to claim 1, wherein determining benchmark equipment having parameter sample values related to the parameter sample values of the equipment under test that match the determined threshold values comprises:

comparing the parameter sample values of the equipment under test that match the determined threshold values with parameters sample values of the benchmark equipment via a K nearest neighbor (KNN) algorithm.

5. The method according to claim 4, wherein when a Euclidean distance between the parameter sample values of the equipment under test that match the determined threshold values and the parameter sample values of the benchmark equipment is below a threshold, the parameter sample values of the benchmark equipment are determined to match the parameter sample values of the equipment under test.

6. The method according to claim 1, wherein the benchmark equipment and the equipment under test are a same type of equipment.

7. The method according to claim 1, further comprising:

receiving, via a user interface, a combination of parameter values for the one or more parameters of the model;

inputting the combination of parameter values into the model;

determining a probability of failure of the equipment under test based on an output value of the model; and

communicating the determined probability to a user interface.

14. The method according to claim 1,

wherein the user interface server is further configured include one or more user controls on the user interface for specifying a combination of parameter values for the one or more parameters;

wherein the analytics processor is further configured to;

input the one or more parameters values into the model;

determine a probability of failure of the equipment under test based on an output value of the model; and

wherein the user interface server is further configured to communicate the determined probability to a user interface.

8. A system for predicting equipment failure, the system comprising:

data reception hardware configured to receive parameter sample values associated with parameters of benchmark equipment and operational status information associated with the benchmark equipment, wherein the parameter sample values and operational status information are periodically acquired;

an analytics hardware module configured to;

generate a model that relates one or more of the parameters to benchmark equipment failure;

for each of the one or more parameters, determine a threshold value at which an output of the model indicates benchmark equipment failure;

determine parameters of an equipment under test for which at least some parameter sample values associated with the parameters match the determined threshold values;

for each determined parameter, determine benchmark equipment having parameter sample values that match the parameter sample values of the equipment under test; and

generate survivability data for the equipment under test based on survivability data associated with the determined benchmark equipment; and

a user interface server configured to receive the generated survivability data from the analytics hardware module; and

generate a user interface that includes a survivability chart to facilitate visualization of the survivability data, and to communicate the user interface to a terminal.

9. The system according to claim 8, wherein the analytics hardware module further comprises a correlation hardware module configured to correlate the parameter sample values associated with parameters of benchmark equipment with the operational status information of benchmark equipment to determine one or more parameters having a greatest relationship to the operational status of the benchmark equipment;

and

wherein the one or more parameters related to the benchmark equipment failure by the model correspond to those parameters having a correlation higher than a predetermined threshold.

10. The system according to claim 8, wherein the generated model corresponds to a linear regression model and the analytics processor further includes model generator hardware that implements the linear regression model.

11. The system according to claim 8, wherein determining benchmark equipment having parameter sample values related to the parameter sample values of the equipment under test that match the determined threshold values comprises:

comparing the parameter sample values of the equipment under test that match the determined threshold values with parameters sample values of the benchmark equipment via a K nearest neighbor (KNN) algorithm.

12. The system according to claim 11, wherein when a Euclidean distance between the parameter sample values of the equipment under test that match the determined threshold values and the parameter sample values of the benchmark equipment is below a threshold, the parameter sample values of the benchmark equipment are determined to match the parameter sample values of the equipment under test.

13. The system according to claim 8, wherein the benchmark equipment and the equipment under test are a same type of equipment.

15. A non-transitory machine-readable storage medium having stored thereon a computer program comprising at least one code section for predicting equipment failure, the at least one code section being executable by a machine for causing the machine to perform acts of:

receiving parameter sample values associated with parameters of benchmark equipment and operational status information associated with the benchmark equipment, wherein the parameter sample values and operational status information are periodically acquired;

generating a model that relates one or more of the parameters to benchmark equipment failure;

for each of the one or more parameters, determining a threshold value at which an output of the model indicates benchmark equipment failure;

determining parameters of an equipment under test for which at least some parameter sample values associated with the parameters match the determined threshold values;

for each determined parameter, determining benchmark equipment having parameter sample values that match the parameter sample values of the equipment under test;

generating survivability data for the equipment under test based on survivability data associated with the determined benchmark equipment; and

communicating the generated survivability data to a user interface.

16. The non-transitory machine-readable storage according to claim 15, further comprising:

correlating the parameter sample values associated with parameters of benchmark equipment with the operational status information of benchmark equipment to determine one or more parameters having a greatest relationship to the operational status of the benchmark equipment; and

wherein the one or more parameters related to the benchmark equipment failure by the model correspond to those parameters having a correlation higher than a predetermined threshold.

17. The non-transitory machine-readable storage according to claim 15, wherein the generated model corresponds to a linear regression model.

18. The non-transitory machine-readable storage according to claim 15, wherein determining benchmark equipment having parameter sample values related to the parameter sample values of the equipment under test that match the determined threshold values comprises:

comparing the parameter sample values of the equipment under test that match the determined threshold values with parameters sample values of the benchmark equipment via a K nearest neighbor (KNN) algorithm.

19. The non-transitory machine-readable storage according to claim 18, wherein when a Euclidean distance between the parameter sample values of the equipment under test that match the determined threshold values and the parameter sample values of the benchmark equipment is below a threshold, the parameter sample values of the benchmark equipment are determined to match the parameter sample values of the equipment under test.

20. The non-transitory machine-readable storage according to claim 15, wherein the benchmark equipment and the equipment under test are a same type of equipment.

21. The non-transitory machine-readable storage according to claim 15, further comprising:

receiving, via a user interface, a combination of parameter values for the one or more parameters of the model;

inputting the combination of parameter values into the model;

determining a probability of failure of the equipment under test based on an output value of the model; and

communicating the determined probability to a user interface.